4551 Background: Surrogate endpoints are needed to shorten the duration of Phase III clinical trials in advanced prostate cancer. Prostate-specific antigen (PSA) is the most studied biomarker in prostate cancer. This study attempts to validate PSA endpoints as surrogates for overall survival (OS). Methods: Individual data from 2161 patients with advanced prostate cancer treated in studies comparing bicalutamide (‘Casodex’), either as monotherapy or in combination with an LHRHa, with castration were used in a meta-analytical approach to surrogate endpoint validation. PSA response, several definitions of time to PSA progression and longitudinal PSA measurements were considered. Results: The analyses confirmed the known association between the PSA endpoints and OS at the individual patient level (biomarker association). However, when comparing patients treated with bicalutamide-based treatment or castration, the effect of hormonal intervention on the PSA endpoint did not predict the effect on OS with a high degree of precision. The association between intervention on any PSA endpoint and OS, measured by the determination coefficient R2 (ranging from 0.10–0.66 for PSA progression, to 0.69 for the whole PSA profile) was generally low. Conclusions: It is a common misconception that a correlation at the individual level between PSA and OS is enough to demonstrate surrogacy. To demonstrate true surrogacy, a high correlation between the treatment effect on the surrogate and the treatment effect on the true endpoint needs to be established across groups of patients treated with two alternative interventions. The level of association observed in this study between the treatment effect on PSA endpoints and that observed on OS was in general low, showing that in Phase III clinical trials of hormonal treatments in advanced prostate cancer, treatment effects on OS cannot be predicted from observed treatment effects on PSA endpoints with a high degree of precision. This study indicates that PSA is unlikely to be a useful surrogate for OS in advanced prostate cancer. ‘Casodex’ is a trademark of the AstraZeneca group of companies. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration AstraZeneca AstraZeneca AstraZeneca AstraZeneca AstraZeneca AstraZeneca AstraZeneca
Read full abstract